The Maf family of proteins are a sub-family of AP-1/CREB/ATF proteins. The first member of the family to be identified, the v-maf oncogene, was originally isolated from a spontaneous musculoaponeurotic fibrosarcoma of chicken and identified as the transforming gene of the avian retrovirus, AS42 (Nishizawa, M. et al. (1989) Proc. Natl. Acad. Sci. USA 86:7711-7715). V-maf encodes a 42 kd basic region/leucine zipper (b-zip) protein with homology to the c-fos and c-jun oncogenes. Its cellular homologue, the c-maf proto-oncogene, which has been isolated from murine cells, has only two structural changes in the coding region from .nu.-maf(Kataoka, K. et al. (1993) J. Virol. 67:2133-2141). The maf family includes c-Maf, mafb, a human retina-specific protein Nrl (Swaroop, A. et al. (1992) Proc. Natl. Acad. Sci. USA 89:266-270), mafk, mafF, mafG and p18. The latter four, mafK, mafF, mafG and p18, each encode proteins that lack the amino terminal two thirds of c-Maf that contains the transactivating domain ("small maf proteins") (Fujiwara, K. T. et al. (1993) Oncogene 8:2371-2380; Igarashi, K. et al. (1995) J. Biol. Chem. 270:7615-7624; Andrews, N. C. et al. (1993) Proc. Natl. Acad. Sci. USA 90:11488-11492; Kataoka, K. et al. (1995) Mol. Cell. Biol. 15:2180-2190).
C-Maf and other Maf family members form homodimers and heterodimers with each other and with Fos and Jun, consistent with the known ability of the AP-1 proteins to pair with each other (Kerppola, T. K. and Curran, T. (1994) Oncogene 9:675-684; Kataoka, K. et al. (1994) Mol. Cell. Biol. 14:700-712). The DNA target sequence to which c-Maf homodimers bind, termed the c-Maf response element (MARE), is a 13 or 14 bp element which contains a core TRE (T-MARE) or CRE (C-MARE) palindrome respectively, c-Maf has been shown to stimulate transcription from the Purkinje neuron-specific promoter L7 (Kurscher, C. and Morgan, J. I. (1994) Mol. Cell. Biol. 15:246-254) and Nrl has been shown to drive expression of the QR1 retina-specific gene (Swaroop, A. et al. (1992) Proc. Natl. Acad. Sci. USA 89:266-270). Additionally, the small mafs have been shown to function as repressors of .alpha. and .beta.-globin transcription when bound as homodimers but are essential as heterodimeric partners with the erythroid-specific factor p45NF-E2 to activate globin gene transcription (Kataoka, K. et al. (1995) Mol. Cell. Biol. 15:2180-2190; Igarashi, K. et al. (1994) Nature 367:568-572). MafK overexpression has been shown to induce erythroleukemia cell differentiation (Igarashi, K. et al. (1995) Proc. Natl. Acad. Sci. USA 92:7445-7449). Moreover, c-Maf has been shown to control the tissue-specific expression of the cytokine interleukin-4 in T helper 2 (Th2) cells (Ho, I-C. et al. (1996) Cell 85:973-983).
The nucleotide sequence of the mouse c-maf proto-oncogene, and predicted amino acid sequence for the mouse c-Maf protein, have been described (Kurscher, C. and Morgan, J. I. (1995) Mol. Cell. Biol. 15:246-254; and Genbank Accession number S74567). The nucleotide sequence of the chicken c-maf proto-oncogene, and predicted amino acid sequence for the chicken c-Maf protein, also have been described (Kataoka et al., Genbank Accession number D28596). However, these non-human c-Maf compositions may not function optimally in human cells and, moreover, use of these compositions in humans is likely to stimulate an immune response, since the chicken or mouse c-Maf would be recognized as "foreign" by the human immune system. Accordingly, there is still a need for human c-Maf compositions that are suitable for use in humans.